development and characterization of microsatellite markers for … tropical/pdf/pubs... ·...

of 15 /15
Development and characterization of microsatellite markers for analysis of population differentiation in the tree legume Acacia koa (Fabaceae: Mimosoideae) in the Hawaiian Islands Rudolph Fredua-Agyeman, Daniel Adamski, Richard Junfu Liao, Clifford Morden, and Dulal Borthakur Abstract: The aim of this research was to develop and use microsatellite markers to characterize the high-value timber tree Acacia koa (koa), which is endemic to the Hawaiian Islands. Genomic DNA fragments of 300–1000 bp were cloned and sequenced following enrichment for microsatellite motifs by PCR using 7 oligonucleotide repeat primers in separate reactions. Among 96 sequences analyzed, 63 contained unique microsatellite motifs flanked by variable sequences. A dual PCR method involving a primer walking step was used to develop 15 primer pairs. Another 16 primer pairs were devel- oped directly from the variable sequences on both sides of the microsatellite motifs. These 31 primer pairs were tested on 172 koa plants representing 11 populations collected from 4 of the major Hawaiian Islands. Nine of the primers that identi- fied polymorphic microsatellite loci and 3 that detected unique alleles exclusively in some populations were used for ge- netic diversity studies of koa. Cluster analysis and multidimensional scaling of the allelic phenotype data revealed that koa from Kauai formed a distinct group separate from koa of the neighboring islands of Oahu, Maui, and Hawaii. The oldest of the four islands, Kauai, also had the most diverse populations of koa. Key words: Acacia koa, dual-suppression PCR, walking sequence and direct sequence strategies, genetic diversity, popula- tion differentiation. Re ´sume ´: Le but de ce travail e ´tait de mettre au point des marqueurs microsatellites pour caracte ´riser l’Acacia koa (koa), une espe `ce forestie `re de grande valeur qui est ende ´mique des ı ˆles hawaı ¨ennes. Des fragments d’ADN ge ´nomique de 300– 1 000 pb ont e ´te ´ clone ´s et se ´quence ´s apre `s enrichissement par PCR pour des microsatellites au moyen de 7 amorces a ` re ´pe ´- titions oligome ´riques dans des re ´actions se ´pare ´es. Parmi les 96 se ´quences analyse ´es, 63 contenaient des microsatellites uni- ques et e ´taient borde ´s de se ´quences variables. Une me ´thode de PCR double incluant une e ´tape de marche chromosomique ae ´te ´ employe ´e pour de ´velopper 15 paires d’amorces. Seize autres paires d’amorces ont e ´te ´ de ´veloppe ´es directement a ` par- tir des se ´quences variables bordant les deux co ˆte ´s des microsatellites. Ces 31 paires d’amorces ont e ´te ´ teste ´es sur 172 plants de koa repre ´sentant onze populations e ´chantillonne ´es sur quatre des principales ı ˆles d’Hawaı ¨. Neuf paires d’amorces qui amplifiaient des locus microsatellites polymorphes et trois paires d’amorces qui amplifiaient des alle `les uniques exclu- sivement chez certaines populations ont e ´te ´ employe ´es pour des e ´tudes de diversite ´ ge ´ne ´tique chez le koa. Des analyses de groupement et d’e ´chelonnement multidimensionnel des donne ´es alle ´liques phe ´notypiques ont re ´ve ´le ´ que le koa de Kauai forme un groupe nettement distinct de ceux des ı ˆles voisines d’Oahu, Maui et Hawaı ¨. Kauai, la plus ancienne des quatre ı ˆles, avaient les populations de koa les plus diverses. Mots-cle ´s : Acacia koa, PCR a ` suppression double, strate ´gies de se ´quence marche ´e et de se ´quence directe, diversite ´ ge ´ne ´- tique et diffe ´renciation des populations. [Traduit par la Re ´daction] Introduction Genetic markers are specific DNA sequences in the ge- nome of an organism that can be used to track inheritance and distribution of parental genotypes among a segregating progeny, and to determine the extent of genetic variation in a population (Schlo ¨tterer 2004). Closely related individuals within a species can have only a limited number of pheno- typic variations, which may also be greatly influenced by changes in the environment (Yadav et al. 2007). On the con- Received 20 March 2008. Accepted 7 September 2008. Published on the NRC Research Press Web site at on 19 November 2008. Corresponding Editor: P. Gustafson. R. Fredua-Agyeman, D. Adamski, R.J. Liao, and D. Borthakur. 1 Department of Molecular Biosciences & Bioengineering, University of Hawaii at Manoa, 1955 East-West Road, Agricultural Science 218, Honolulu, HI 96822, USA. C. Morden. Department of Botany, University of Hawaii at Manoa, Honolulu, HI 96822, USA. 1 Corresponding author (e-mail: [email protected]). 1001 Genome 51: 1001–1015 (2008) doi:10.1139/G08-087 # 2008 NRC Canada

Author: others

Post on 22-Jul-2020




0 download

Embed Size (px)


  • Development and characterization ofmicrosatellite markers for analysis of populationdifferentiation in the tree legume Acacia koa(Fabaceae: Mimosoideae) in the Hawaiian Islands

    Rudolph Fredua-Agyeman, Daniel Adamski, Richard Junfu Liao, Clifford Morden,and Dulal Borthakur

    Abstract: The aim of this research was to develop and use microsatellite markers to characterize the high-value timbertree Acacia koa (koa), which is endemic to the Hawaiian Islands. Genomic DNA fragments of 300–1000 bp were clonedand sequenced following enrichment for microsatellite motifs by PCR using 7 oligonucleotide repeat primers in separatereactions. Among 96 sequences analyzed, 63 contained unique microsatellite motifs flanked by variable sequences. A dualPCR method involving a primer walking step was used to develop 15 primer pairs. Another 16 primer pairs were devel-oped directly from the variable sequences on both sides of the microsatellite motifs. These 31 primer pairs were tested on172 koa plants representing 11 populations collected from 4 of the major Hawaiian Islands. Nine of the primers that identi-fied polymorphic microsatellite loci and 3 that detected unique alleles exclusively in some populations were used for ge-netic diversity studies of koa. Cluster analysis and multidimensional scaling of the allelic phenotype data revealed that koafrom Kauai formed a distinct group separate from koa of the neighboring islands of Oahu, Maui, and Hawaii. The oldestof the four islands, Kauai, also had the most diverse populations of koa.

    Key words: Acacia koa, dual-suppression PCR, walking sequence and direct sequence strategies, genetic diversity, popula-tion differentiation.

    Résumé : Le but de ce travail était de mettre au point des marqueurs microsatellites pour caractériser l’Acacia koa (koa),une espèce forestière de grande valeur qui est endémique des ı̂les hawaı̈ennes. Des fragments d’ADN génomique de 300–1 000 pb ont été clonés et séquencés après enrichissement par PCR pour des microsatellites au moyen de 7 amorces à répé-titions oligomériques dans des réactions séparées. Parmi les 96 séquences analysées, 63 contenaient des microsatellites uni-ques et étaient bordés de séquences variables. Une méthode de PCR double incluant une étape de marche chromosomiquea été employée pour développer 15 paires d’amorces. Seize autres paires d’amorces ont été développées directement à par-tir des séquences variables bordant les deux côtés des microsatellites. Ces 31 paires d’amorces ont été testées sur 172plants de koa représentant onze populations échantillonnées sur quatre des principales ı̂les d’Hawaı̈. Neuf paires d’amorcesqui amplifiaient des locus microsatellites polymorphes et trois paires d’amorces qui amplifiaient des allèles uniques exclu-sivement chez certaines populations ont été employées pour des études de diversité génétique chez le koa. Des analyses degroupement et d’échelonnement multidimensionnel des données alléliques phénotypiques ont révélé que le koa de Kauaiforme un groupe nettement distinct de ceux des ı̂les voisines d’Oahu, Maui et Hawaı̈. Kauai, la plus ancienne des quatreı̂les, avaient les populations de koa les plus diverses.

    Mots-clés : Acacia koa, PCR à suppression double, stratégies de séquence marchée et de séquence directe, diversité géné-tique et différenciation des populations.

    [Traduit par la Rédaction]


    Genetic markers are specific DNA sequences in the ge-nome of an organism that can be used to track inheritanceand distribution of parental genotypes among a segregating

    progeny, and to determine the extent of genetic variation ina population (Schlötterer 2004). Closely related individualswithin a species can have only a limited number of pheno-typic variations, which may also be greatly influenced bychanges in the environment (Yadav et al. 2007). On the con-

    Received 20 March 2008. Accepted 7 September 2008. Published on the NRC Research Press Web site at on19 November 2008.

    Corresponding Editor: P. Gustafson.

    R. Fredua-Agyeman, D. Adamski, R.J. Liao, and D. Borthakur.1 Department of Molecular Biosciences & Bioengineering, Universityof Hawaii at Manoa, 1955 East-West Road, Agricultural Science 218, Honolulu, HI 96822, USA.C. Morden. Department of Botany, University of Hawaii at Manoa, Honolulu, HI 96822, USA.1Corresponding author (e-mail: [email protected]).


    Genome 51: 1001–1015 (2008) doi:10.1139/G08-087 # 2008 NRC Canada

  • trary, genetic markers based on DNA sequences can beinnumerable and are generally not subject to environmentalinfluences (Manifesto et al. 2001). However, because of thevastness of genome sequences, analyses of random genesequences between individuals may not always lead todetectable differences. There are certain highly variableregions in the genomes of organisms where differences insequences among individuals are more prevalent (Ellegren2004; Weber 1990). Numerous methods of easy determina-tion of differences within these hotspots of genetic variation,such as restriction fragment length polymorphism (RFLP),randomly amplified polymorphic DNA (RAPD), amplifiedfragment length polymorphism (AFLP), single nucleotidepolymorphism (SNP), and DNA repeat variation analyses,which include microsatellites, have been already describedin other literature (Skolnick and White 1982; Tautz 1989;Williams et al. 1990; Vos et al. 1995; Weising et al. 1995;Brookes 1999; Toth et al. 2000). Molecular markers identi-fied from these variable regions have wide application in avariety of fields including phylogeny, taxonomy, ecology,genetics, and plant and animal breeding (McCouch et al.1997; Ouborg et al. 1999; Luo et al. 2001; Bussell et al.2005; Wang et al. 2006). These markers can also provideinformation about allelic variation at a given locus that maybe useful for understanding microevolution of populations atvarying geographical and environmental locations (Zwettleret al. 2002).

    Microsatellites, or simple sequence repeats (SSR), arepresently among the most widely used DNA markers fordetection of genetic variation in different organisms (Selkoeand Toonen 2006). Microsatellite markers, first developedfor use in genetic fingerprinting of humans, are composedof tandem repeats of 1–6 nucleotides and are dispersedthroughout the genome of an organism (Tautz and Renz1984; Litt and Luty 1989). Microsatellite markers havebeen successfully applied in population genetics studies ofpolyploid trees including Acacia species (Butcher et al.2000; Otero-Arnaiz et al. 2005; Millar and Byrne 2007;Ruiz-Guajardo et al. 2007). Acacia koa (koa) is a memberof the subgenus Phyllodineae (Fabaceae: Mimosoideae) andis a tetraploid (2n = 52) species that is endemic to theHawaiian Islands (Atchison 1948; Carr 1978). It is an eco-system-dominant tree in the forests of Hawaii and is asource for high-value timber. It is also considered to be animportant canopy tree that provides watershed and habitatfor other endemic species of Hawaii (St. John 1979; White-sell 1990). In natural forests, large phenotypic differencesamong island types of koa are observed in seed size andshape, pubescence, phyllode width and curvature, floralmorphology and pigmentation, retention of juvenile leaves,and branch bark color (Sun et al. 1996; Daehler et al.1999). Microsatellite markers for studying genetic diversityin koa had not yet been developed. Here, microsatellitemarkers were developed and used for the first time to char-acterize distinct morphological forms of koa among fourmajor Hawaiian Islands.

    Materials and methods

    Plant materials and DNA extractionFor microsatellite isolation, total genomic DNA was ex-

    tracted from a koa leaf sample obtained from Island of Ha-waii using the DNeasy Plant Mini Kit (QIAGEN Inc.,Valencia, California, USA) according to the manufacturer’sinstructions. The quality of the DNA was evaluated by spec-trometry using the 260/280 nm absorbance ratio method andthe DNA concentration was estimated at 260 nm with anND-1000 Nanodrop spectrophotometer (Nanodrop Technol-ogies, Inc., Wilmington, Delaware, USA). For testing micro-satellite primers, total genomic DNA from 172 koa treesrepresenting 11 populations from the islands of Kauai,Oahu, Maui, and Hawaii (Table 1) was obtained from theHawaiian Plant DNA Library (Morden et al. 1996; Randelland Morden 1999).

    Restriction fragmentation, size selection, andoligonucleotide repeat PCR

    The koa genomic DNA was separately digested with5 blunt-end-producing restriction enzymes (RsaI, AluI, ScaI,HaeIII, and EcoRV), and fragments of 300–1000 bp wereextracted from agarose gels using the QIAquick Gel Extrac-tion Kit (QIAGEN Inc.). The DNA digests were pooled tocreate a library of overlapping fragments. Fragments thatcontained 2 or more microsatellite motifs were amplifiedfrom the DNA library using 7 oligonucleotide repeats([AC]10, [AAC]8, [AG]10, [AAG]8, [TG]10, [AAT]8, and[TAGA]8) in separate reactions. All oligonucleotide repeatsfor this study were obtained from Integrated DNA Technol-ogies (San Diego, California, USA). PCR was performed ina total volume of 50 mL containing 3–10 ng of genomicDNA, 1� Taq buffer, 2.5 mmol/L MgCl2, 200 mmol/L eachdNTP, 0.2 mmol/L primer, and 1.25 U of Taq polymerase(Promega Corp., Madison, Wisconsin, USA). The reactionswere carried out in an MJ Research PTC-200 DNA EngineThermal Cycler (Bio-Rad Laboratories, Hercules, California,USA); amplifications consisted of an initial denaturationstep of 5 min at 94 8C, 35 cycles of 1 min at 94 8C, 1 minat 54 8C, and 1 min 30 s at 72 8C, and a final extension of30 min at 72 8C. Aliquots of the PCR products were electro-phoresed on 2% agarose gels, stained for 30 min in ethidiumbromide, and photographed under ultraviolet light for visual-ization of the amplified fragments.

    Construction of genomic DNA libraries and sequencingof amplified fragments

    The amplified fragments were directly ligated into thepGEM-T Easy Vector (Promega Corp.) and ligation reac-tions were used to transform competent E. coli DH5a cells.The cloned fragments from the libraries were evaluated bycolony PCR using the M13 forward and reverse primers.For each library, 10 to 30 transformants containingdifferent-sized fragments were grown overnight in liquidLB medium containing 100 mg/mL ampicillin. PlasmidDNA was extracted from positive clones using the QIAprepSpin Miniprep Kit (QIAGEN Inc.) and then used as templatein the sequencing reactions. Sequencing reactions were per-formed using the ABI PRISM BigDye Terminator v3.1 CycleSequencing Kit and analyzed on an ABI PRISM 3730xl DNAanalyzer (Applied Biosystems, Foster City, California, USA)at the Advanced Studies in Genomics, Proteomics and Bio-informatics sequencing facility at the University of Hawaiiat Manoa. Clone sequences were extracted from the chroma-

    1002 Genome Vol. 51, 2008

    # 2008 NRC Canada

  • tograms using the SeqVerter program (GeneStudio, Inc., Su-wannee, Georgia, USA). To eliminate redundant clones, theretrieved sequences were aligned using the ClustalW pack-age at the European Bioinformatics Institute (EBI; The isolated sequences were comparedagainst the GenBank database sequences available at the Na-tional Center for Biotechnology Information (NCBI) usingthe BLASTN search program ( The sequences obtained in this study have beendeposited in the GenBank database under the accessionnumbers EU574702–EU574732.

    Genomic SSR analyses and primer developmentTwo approaches, namely, ‘‘walking sequence’’ and

    ‘‘direct sequence’’, were used for primer development(Fig. 1). For the ‘‘walking sequence’’ approach, genomic se-quences that contained microsatellite motifs only at the endswere selected for amplification using two primers (Lian etal. 2001; Siebert et al. 1995). One of the primers, known asthe ‘‘downstream primer’’, was developed from the variablesequence located adjacent to the 3’ end of the microsatellitewithin the cloned fragment, using the Primer3 program(Rozen and Skaletsky 2000). For developing the ‘‘upstreamprimer’’, which is based on variable sequence on the originalgenomic DNA fragment containing the same microsatellitemotif, a multi-step approach was taken as follows. (i) Koagenomic DNA was digested with 5 blunt-end-producing re-striction enzymes (RsaI, AluI, ScaI, HaeIII, and EcoRV)separately and the resulting fragments were purified usingthe QIAquick PCR Purification Kit. (ii) The digested ge-nomic DNA was ligated at both ends to the blunt end of anadapter obtained by annealing a 48 nucleotide (nt) primerwith an 11 nt primer; this adapter contains an amino groupat the 3’ end that blocks further extension, thereby leavingthe longer adapter primer free (Table 2). The short primerwas complementary to the 3’ end of the long primer, produc-ing a blunt end and a 5’ overhang of 37 nt. (iii) The adapter-ligated DNA fragments were used as templates in PCRusing the downstream primer and a primer containing a 27-mer direct sequence of the 5’ overhang of the adapter. Thenewly synthesized single-stranded DNA from the down-stream primer serves as the template for annealing of theadapter primer. In this way, the microsatellite and adjacentregions were amplified. (iv) PCR products were cloned and

    sequenced. If multiple bands were obtained from the PCRamplification, a nested primer approach was applied. A sec-ond downstream primer was developed from sequences adja-cent to the first downstream primer and used with theadapter primer to obtain a discrete PCR product, which wascloned and sequenced. (v) Upstream primers were developedfrom the non-repeated variable sequence adjacent to theadapter sequence. For the ‘‘direct sequence’’ primer develop-ment strategy, forward and reverse primers were developedfrom the sequenced fragments that contained centrally lo-cated microsatellite motifs in addition to variable sequencesof at least 60 bp on both sides of the motifs.

    Testing primers by PCR amplification of genomic DNAPCR was performed in a total volume of 20 mL contain-

    ing 3–10 ng of genomic DNA, 1� Taq buffer, 2.5 mmol/LMgCl2, 200 mmol/L each dNTP, 0.2 mmol/L primer, and1.25 U of Taq polymerase (Promega Corp.). Each micro-satellite marker was tested separately with genomic DNAfrom each of the 172 test plants. The reactions were carriedout using the same PCR amplification protocol describedabove. Aliquots of the PCR products from the samples wereelectrophoresed on 2% agarose gels, stained for 30 min inethidium bromide, and photographed under ultraviolet lightfor visualization to confirm the success of amplification re-actions. One primer from each pair that positively amplifiedgenomic DNA to produce a single band was fluorescentlylabeled with FAM, VIC, PET, or NED (Applied Bio-systems).

    Allele size determination and polymorphism detectionAliquots (1 mL) of the amplified DNA fragments were

    size-separated by capillary electrophoresis on an ABI PRISM3730xl DNA analyzer (Applied Biosystems) at the Ad-vanced Studies in Genomics, Proteomics and Bioinformaticssequencing facility at the University of Hawaii at Manoa.The amplified DNA fragment lengths or allele sizes weredetermined by comparison with the standard size markerGS 500 ROX (Applied Biosystems) using the GeneMarkersoftware program, version 1.60 (SoftGenetics, LLC, StateCollege, Pennsylvania, USA). To confirm the reproducibilityof peaks and the accuracy of size calling, PCR amplifica-tions were repeated on about 10% of the samples and allelesizes for each microsatellite marker were rescored.

    Table 1. List of 172 Acacia koa trees analyzed by microsatellite markers.

    Sample IDs HPDL acc. Nos.No. ofsamples

    Island oforigin Population sampled

    H1BPH–H16BPH 2005–2020 16 Hawaii Bird Park, Hawaii Volcanoes National ParkH17LNR–H32LNR 2021–2036 16 Hawaii Laupahoehoe Natural Area ReserveM33MRW–M48MRW 2379–2394 16 Maui Mahana Ridge WestM49MFR–M64MFR 2395–2410 16 Maui Makawao Forest ReserveM65KFR–M80KFR 2411–2426 16 Maui Kapunakea Forest ReserveK81NPC–K96NPC 1944–1959 16 Kauai Na Pali CoastK97MKR–K110MKR 1980–1993 14 Kauai Makaha Ridge, KokeeK111HAL–K126HAL 2979–2994 16 Kauai HanaleiO127KHV–O142KHV 1887–1902 16 Oahu Kahana ValleyO143HLR–O157HLR 1871– ! 1886 15 Oahu Hawaii Loa RidgeO158WNK–O172WNK 1733– ! 1748 15 Oahu Waianae Kai

    Note: Source of DNA was the Hawaiian Plant DNA Library (HPDL). ! indicates numbering break in HPDL accessions.

    Fredua-Agyeman et al. 1003

    # 2008 NRC Canada

  • Microsatellite allele counting using peaks of electrophero-grams was used for the determination of the presence of anallele at a locus. The number of alleles per locus was deter-mined by counting the total number of alleles for each locusfollowing genotyping of 172 koa plants with the correspond-ing set of primers. The polymorphic information content(PIC) of the primers was calculated according to the equa-tion of Anderson et al. (1993):

    ½1� PIC ¼ 1�Xn


    where n is the total number of alleles detected for each pri-mer pair and Pi is the frequency of the ith allele detected.

    Markers that identify polymorphic microsatellite loci werethen used for studying genetic diversity of koa.

    Microsatellite data analysisAs has been previously found with polyploids (De Silva

    et al. 2005), complexity of inheritance patterns and the pres-ence of multiple alleles in koa makes it difficult to assignallele dosage or the copy number of each allele at a specificlocus. A molecular binary phenotype was therefore used toassign microsatellite DNA genotypes for individual plants,in which amplified fragments were analyzed as dominant–recessive markers. The presence of an allele was scored as‘‘1’’ and its absence as ‘‘0’’. Alleles were scored for 12markers (see Results) following capillary electrophoresis of

    Fig. 1. (a) Steps in cloning microsatellite motifs: (1) digestion of genomic DNA to produce blunt ends, (2) isolation of fragments of 300–1000 bp, (3) some of the purified fragments may contain microsatellite motifs, (4) enrichment of microsatellite motifs by oligonucleotiderepeat PCR, (5) fragments containing microsatellite motifs at the two ends were used for primer development using the walking sequencestrategy, in which a downstream primer is made from the variable sequence in the middle. (b) Additional steps in the walking sequencestrategy: (6) adapter is ligated to the blunt-ended genomic fragments, (7) microsatellite motifs with upstream variable regions are amplifiedby nested PCR using adapter-specific primers and downstream primers, and (8) the upstream primer, developed from the variable upstreamregion, is used with the downstream primer to amplify the microsatellite locus.

    Table 2. Adapter and primer sequences used for the development of microsatellite markers by the dual-suppression PCR method.

    Oligonucleotide Sequence

    Single-stranded adapter sequences48 nt 5’-GTAATACGACTCACTATAGGGCACGCGTGGTCGACGGCCCGGGCTGGT-3’11 nt 5’-ACCAGCCCGGG-NH2-3’

    Overhang adapter derived primer sequences used for nested PCR27 nt 5’-CCATCGTAATACGACTCACTATAGGGC-3’18 nt 5’-CTATAGGGCACGCGTGGT-3’

    1004 Genome Vol. 51, 2008

    # 2008 NRC Canada

  • the 172 koa samples. The data, organized in a spreadsheet,were then used for further analysis to determine geneticsimilarities between different island types of koa.

    Cluster analysis and multidimensional scaling of allelicphenotype data

    Allelic phenotype data for the 172 plants representing 11populations of koa were analyzed using the SIMQUAL(similarity for qualitative data) module to compute pairwisesimilarity coefficients (Jaccard 1908; Dice 1945). The simi-larity estimates were used to draw UPGMA (unweightedpair group method with arithmetic means) dendrogramswith the SAHN (sequential, agglomerative, hierarchical andnested clustering) module of the NTSYSpc 2.2 program (Ex-eter Software Co., Setauket, New York, USA). Bootstrapanalysis was performed with 1000 repetitive samplings ofthe data and a consensus tree was constructed. Thecophenetic correlation coefficient was calculated to test thegoodness of fit between the original similarity matrices andthe cophenetic matrices using a Mantel test (Mantel 1967)and the COPH and MXCOMP modules of the NTSYSpc2.2 program. High values indicate that the phenograms rep-resent the relationships among the populations. The principalcoordinate analyses (PCO) were also performed on the samesimilarity indices using the NTSYSpc 2.2 program to pro-vide a pictorial representation of relationships among popu-lations. The first 3 principal axes were used to plot scatterdiagrams that show the relationships among the different is-land types.

    Measuring genetic diversity and differentiation statisticsCytogenetic investigation of koa revealed a maximum of

    one putative pair of chromosomes with secondary constric-tions, which is consistent with the ploidy level of an allote-traploid species (Langer and Koul 1983; Coulaud et al.1995; Shi 2003). Therefore, in addition to the above analy-sis, we estimated phenotype-based genetic diversity and dif-ferentiation statistics with the FDASH program developedfor use in allopolyploids by Obbard et al. (2006b). Thephenotype-based statistics performed extremely well undercomputer simulations and were recently used in populationgenetics studies of the allohexaploid herb Geum triflorum(Hamilton and Eckert 2007) and hexaploid populations ofMercurialis annua (Obbard et al. 2006a).

    The overall genetic diversity across all populations wasdetermined by estimating H 0T, a parameter that is based onthe average number of unshared alleles between pairs of in-dividuals taken from all populations (Obbard et al. 2006b):

    ½2� H 0T ¼1

    nðn� 1ÞXn






    where H 0T is the average number of unshared alleles be-tween pairs of individuals across all populations, n is the to-tal number of individuals, and xijk = 1 if only one of theindividuals i and j carry allele k, otherwise xijk = 0. For mea-suring genetic diversity within a population, H 0S was calcu-lated based on the average number of unshared allelesbetween pairs of individuals taken from the same popula-tion.

    The genetic differentiation among populations was inves-

    tigated by estimating F 0ST, a parameter that is based on theproportion of total diversity found between populations (Ob-bard et al. 2006b):

    ½3� F0ST ¼H 0T � H 0S

    H 0T

    where F 0ST is the genetic variance among populations rela-tive to the total variation and H 0S is the mean within-popula-tion diversity. The statistical significance of H 0T, H

    0S, and

    F 0ST was tested after 1000 bootstraps with 95% nominal con-fidence intervals.


    Oligonucleotide repeat PCR analyses of genomicfragments

    Restriction digestion with each of the 5 blunt-end-producing restriction enzymes resulted in smeared bandingpatterns, indicating generation of numerous fragments ofsizes less than 3.0 kb. When 7 oligonucleotide repeatswere used separately as PCR primers for amplification ofthe gel-extracted genomic DNA fragments, only 4 primersresulted in numerous amplified bands. Positive amplifica-tions by these primers indicated that genomic DNA frag-ments containing AC, AG, TG, and AAG repeats wereselectively amplified (Fig. 2). These oligonucleotide pri-mers were expected to amplify internal segments of thedigested DNA fragments between two microsatellite motifs.The resulting PCR products did not contain sequencesbeyond the two motifs on the ends of the template. Fourclone libraries were constructed with the amplified ge-nomic DNA fragments containing the repeated sequencesand these were designated as AC, AG, TG, and AAGlibraries. Several hundred colonies were obtained for eachlibrary. Representative clones from the four librariesshowed cloned fragments in the size range of 300 to1000 bp (Fig. 3).

    Sequence analyses of cloned AC-, AG-, TG-, and AAG-rich fragments

    Among 511 randomly selected colonies from the fourlibraries that were screened by colony PCR using M13 pri-mers for amplified fragments of different sizes, a total of96 clones were selected as representatives of inserts of dif-ferent sizes for sequence analyses (Table 3). Among these,63 clones (66%) had unique microsatellite sequences. Theremaining 33 clones displayed the same sequences presentin the selected unique clones. Because single oligonucleo-tide repeat primers were used for selective amplification ofmicrosatellite-rich DNA fragments, all clones contained re-verse complementary repeated motifs at the two ends. Theclones containing AC, AG, AAG, and TG motifs at the 5’ends carried GT, CT, CTT, and CA motifs, respectively, atthe 3’ ends. The average size of the inserts obtained fromthe walking sequence method was 232 bp, compared with199 bp inserts obtained from the direct sequence method.Among the selected clones, 43% contained centrally locatedmicrosatellite motifs in addition to those at the ends. Fifteendifferent repeats were identified in the central motifs, amongwhich 2 were similar to the terminal motifs identified by thewalking sequence strategy. The central motifs were also

    Fredua-Agyeman et al. 1005

    # 2008 NRC Canada

  • shorter than the motifs at the two ends. Eighty percent of themicrosatellites, terminal or central, contained perfectdinucleotide or trinucleotide repeats. About 13% of themicrosatellites contained two adjacent distinctive repeats,while about 7% had 2–4 bp intervening sequences withinthe motifs. As expected, almost all clones present in the(AC)10 library were also present in the complementary(TG)10 library.

    Primer developmentPrimer development was not possible for 26 of the 63 se-

    lected sequences because the flanking regions containedsimilar microsatellite-like sequences or were too short forprimer development. A total of 37 sequences from the fourlibraries were found to be suitable for primer development.Twenty-one sequences that contained microsatellite motifsat the ends were selected for primer development using the‘‘walking sequence’’ strategy. The remaining 16 sequencesthat contained microsatellite motifs in the middle and hadvariable flanking regions of at least 60 bp on both sideswere selected for ‘‘direct sequence’’ primer development.

    With the ‘‘walking sequence’’ strategy, 16 microsatellite-containing regions were successfully amplified, appearingas discrete single bands on agarose gels. The amplifiedproducts from the other 5 sets of primers showed multiplebands or smears and were not used for further analyses. ThePCR products from the 16 reactions that showed discretesingle bands were cloned and sequenced. The sequence datashowed that the amplified fragments contained microsatellite

    loci and also the adjacent 3’ regions beyond the loci. Fromthe variable sequence between the adapter sequence and themicrosatellites, 16 upstream primers were developed thathad the same orientation as the adapter primers and oppositedirectionality from the downstream primers. Although these16 upstream primers and their corresponding downstreamprimers (Table 4) were expected to amplify 16 differentmicrosatellite loci, only 15 of these primer pairs successfullyamplified DNA fragments from test populations under opti-mized reaction conditions.

    An additional 16 primer pairs were developed by usingthe ‘‘direct sequence’’ primer development strategy from thesequenced fragments that contained centrally located micro-satellite motifs in addition to variable sequences of at least60 bp on both sides of the motifs. The variable regions onboth sides of the microsatellite motifs were used for devel-oping a pair of forward and reverse primers for each of the16 sequences. These 16 primer pairs could amplify theexpected microsatellite loci of the test populations underoptimized reaction conditions (Table 5).

    Polymorphism analysesAll 31 primer pairs could amplify microsatellite loci of

    the 172 koa test plants. Characterization of each micro-satellite locus was based on the number of alleles detectedon electropherograms obtained from capillary electrophore-sis of the PCR products (Fig. 4). The number of alleles perlocus and the polymorphic information content (PIC) of theprimers, based on peak counts in the electropherograms forthe test population, are presented in Tables 4 and 5. Thehighest number of alleles per locus was obtained for Ak08,followed by Ak284 and Ak99. None of the microsatelliteloci was found to be monomorphic. The microsatellite lociamplified by 4 of the primer pairs were dimorphic, whilethose amplified by 4 other primer pairs were trimorphic.The remaining 23 pairs of primers amplified microsatelliteloci that had 4 or more alleles. The PIC values ranged from0.13 to 0.89 and did not correlate with the number of allelesdetected at a particular locus.

    Microsatellite marker selection and allele frequencydetermination

    Nine highly variable koa-specific microsatellite markersthat were part of the initial set of 31 markers were selectedfor genetic diversity studies. Three additional primers that

    Fig. 2. Electrophoretic profiles showing amplified PCR productsobtained from four of the seven oligonucleotide repeat primers.Lanes for oligonucleotides (AC)10, (AG)10, (TG)10, and (AAG)8showed PCR products, while those for (AAC)8, (AAT)8, and(TAGA)8 showed faint or no products. M, DNA size marker.

    Fig. 3. PCR amplification of plasmid DNA of positive clones withM13 primers indicated that fragments in the size range of 300–1000bp were amplified. The first lane shows the DNA size marker.

    Table 3. Summary of the clone library constructed with seven oli-gonucleotide repeats.

    Number of clones


    Screenedby colonyPCR

    Culturedfor plas-mid DNA

    Containing uniquemicrosatellitemotif

    (AC)10 133 30 19(AG)10 255 30 27(TG)10 16 16 3(AAG)8 107 20 14(AAC)8 3 — —(AAT)8 2 — —(TAGA)8 4 — —Total 520 96 63

    1006 Genome Vol. 51, 2008

    # 2008 NRC Canada

  • Table 4. Primer sequences and allele characteristics of 15 ‘‘downstream’’ and ‘‘upstream’’ microsatellite markers isolated from Acacia koa by the ‘‘walking sequence’’ strategy.

    No. Locus LibraryDownstream primer (5’–3’)Upstream primer (5’–3’) Repeat motif Ta (8C) Dye

    Allele size(bp)

    No. ofalleles PIC

    GenBankacc. No.























  • Table 5. Primer sequences and allele characteristics of 16 forward and reverse microsatellite markers isolated from Acacia koa by the ‘‘direct sequence’’ strategy.

    No. Locus LibraryForward primer (5’–3’)Reverse primer (5’–3’) Repeat motif Ta (8C) Dye

    Allele size(bp)

    No. ofalleles PIC

    GenBankacc. No.

























  • were observed to detect unique alleles exclusively in somepopulations were also included in the analysis. A total of116 distinct alleles were identified after 172 koa trees weregenotyped with the 12 selected markers. The average num-ber of alleles per locus was 9.7. A summary of the loci ana-lyzed and the number of alleles detected in each populationis presented in Table 6. The highest number of alleles (62)was detected in the 16 samples from the Hanalei population(HAL) on Kauai, while the lowest number of alleles (39)was detected in 16 samples from the Bird Park (BPH) popu-lation on Island of Hawaii.

    Genetic relationships among populations based onsimilarities and principal coordinate analyses

    The binary data scores obtained from genotyping 172 koatrees with the 12 selected markers were used to generateDice’s and Jaccard’s coefficients. Both Dice’s and Jaccard’scoefficients avoid shared absences, although Dice’s co-efficient gives more weight to matches, resulting in highersimilarity values compared with those from Jaccard’s co-

    efficient. Regardless of which similarity coefficient wasemployed, the UPGMA cluster analysis of the 172 koa treesreadily separated the populations into two major clusters, Aand B (Fig. 5), with minor differences in relationshipswithin the major clusters (Figs. S1, S2, and S3).2 Cluster Acan be subdivided into 3 minor subgroups, G1, G2, and G3,at the demarcation having 64% and 47% similarities forDice’s and Jaccard’s coefficients, respectively. In bothUPGMA dendrograms, subgroup G1 contained 75% of thekoa populations from Island of Hawaii, of which 62.5% and37.5% were LNR and BPH genotypes, respectively. Only3% of the samples in this subgroup were from another is-land, Maui. This represented 2% of koa populations fromMaui. Subgroup G2 contained the remaining 25% of thekoa populations from Island of Hawaii, 77% of koa popula-tions from Maui, and 2.6% of koa populations from Oahu.The Island of Hawaii genotypes in this subgroup were 6.3%LNR and 93.7% BPH. The Maui genotypes were 40.6%MFR, 43.2% MRW, and 16.2% KFR. Subgroup G3 con-tained 14.6% of the remaining 21% of the koa populations

    Fig. 4. Electropherogram showing all 4 alleles detected at locus Ak16. The x-axis shows fragment size (bp) and the y-axis shows the peakheight.

    2 Supplementary data for this article are available on the journal Web site ( or may be purchased from the Depositoryof Unpublished Data, Document Delivery, CISTI, National Research Council Canada, Building M-55, 1200 Montreal Road, Ottawa, ONK1A 0R6, Canada. DUD 3840. For more information on obtaining material refer to

    Fredua-Agyeman et al. 1009

    # 2008 NRC Canada

  • from Maui, 97.8% of koa populations from Oahu, and 2.2%of koa populations from Kauai. The Oahu genotypes were31.1% HLR, 35.6% KHV, and 33.3% WNK. All the KHVand WNK samples analyzed were present in this subgroup.Also, 93.3% of HLR genotypes were present in this sub-group. Cluster B, or subgroup G4, contained the remaining6.3% of koa populations from Maui and 97.8% of koa pop-

    ulations from Kauai. Thus, UPGMA cluster analysis of the172 koa trees was able to segregate the 11 populations into4 clusters, G1, G2, G3, and G4, which contained 75%, 77%,97.8%, and 97.8% of the samples from Hawaii, Maui, Oahu,and Kauai, respectively.

    The binary scores for the presence or absence of a peak inthe electropherogram were also used for PCO by both

    Table 6. Number of alleles detected at each of 12 microsatellite loci in 11 populations of Acacia koa.

    Number of alleles

    Population Ak05 Ak08 Ak10 Ak15 Ak16 Ak36 Ak41 Ak44 Ak89 Ak99 Ak196 Ak284 TotalBPH 2 4 3 2 1 2 2 4 2 8 1 8 39LNR 4 5 4 2 1 3 3 3 5 5 3 11 49MRW 4 4 5 3 2 5 4 3 2 4 1 5 42MFR 4 4 4 3 1 4 3 5 2 6 3 7 46KFR 3 4 3 3 2 5 4 8 4 6 2 10 54NPC 3 8 2 2 2 3 4 5 2 5 2 9 47MKR 3 11 4 3 3 3 3 6 2 8 2 10 58HAL 4 12 2 2 2 2 3 7 3 9 2 14 62KHV 4 5 4 2 1 5 4 4 4 6 1 4 44HLR 5 5 4 2 1 5 3 3 3 6 2 6 45WNK 3 6 3 2 2 4 4 3 4 5 2 4 42Total 6 17 8 4 4 8 8 13 11 14 7 16 116

    Fig. 5. UPGMA dendrogram derived from Dice’s coefficient of similarity, demonstrating the genetic relationships among 172 individualtrees of Acacia koa based on 12 microsatellite markers. The 11 populations were grouped into two major clusters, A and B (subgroup G4),with cluster A further divided into 3 subgroups (G1, G2, and G3). Cophenetic correlation coefficient = 0.92.

    1010 Genome Vol. 51, 2008

    # 2008 NRC Canada

  • Dice’s and Jaccard’s methods. To capture the maximum var-iation in our microsatellite diversity data, we used the first,second, and third principal axes, with a total variability of54%. Plots derived from the first 3 principal coordinateaxes using either method showed that koa populations fromKauai form a distinct group separate from koa populationson all the other neighboring islands (Fig. 6). Thus, PCObased on both Dice’s and Jaccard’s coefficients was able todivide koa populations from the four islands into two majorgroups.

    Genetic diversity statisticsThe overall genetic diversity (H 0T) estimated across all koa

    populations with the FDASH program was 1.001. The aver-age within-population genetic diversity (H 0S) ranged from0.566 to 0.988, indicating that the within-population geneticdiversity accounts for a very high percentage of the overalldiversity. This trend was due to significantly higher H 0Swithin populations from all four islands (randomization test,P < 0.05). Pairwise genetic differentiation, determined byF0ST, ranged from 0.028 to 0.230 and averaged 0.093, 0.060,0.065, and 0.070 for Kauai, Oahu, Maui, and Hawaii popu-lations, respectively (Table 7). These averages were not sig-nificantly different from each other, suggesting thatdifferentiation of koa populations within islands is very sim-ilar across the four islands sampled (t test, P ‡ 0.38). Pair-wise comparison of the 11 populations showed that koafrom Maui Makawao Forest Reserve (MFR) were closely re-lated to those from Kapunakea Forest Reserve (Table 7).Similarly, the Oahu Hawaii Loa Ridge (HLR) populationwas closely related to the Waianae Kai (WNK) population.On the other hand, the population from Mahana Ridge West

    (MRW) on Maui was most dissimilar to the population fromWaianae Kai (WNK) on Oahu. Thus, the spatial separationof the islands also appeared to influence genetic differentia-tion of koa populations.


    Microsatellite loci of plants and animals are known to beassociated mostly with noncoding regions or regions flank-ing coding sequences of the genome, and are heritable asMendelian traits (Hancock 1995; Toth et al. 2000; Ellegren2004; Schlotterer 2004). These loci are used as markers inlinkage studies, for construction of genetic maps, and fordetermining genetic diversity in populations. Isolation ofmicrosatellite markers is especially useful in species forwhich sufficient genomic sequences are not available (Zaneet al. 2002). We used the dual-suppression PCR protocol(Lian et al. 2001) for enrichment and isolation of adjacentmicrosatellite loci from genomic DNA of koa. As a result,nearly all the cloned fragments contained microsatellite se-quences. Also, during the second PCR amplification usingthe downstream and adapter primers, we selected fragmentsof different sizes for cloning and sequencing. This resultedin the isolation of mostly unique microsatellite loci andonly a few redundant sequences. Analyses of the sequencesof the PCR fragments obtained from the initial step of thisprotocol showed that about 18% of the fragments containedmicrosatellite motifs in the middle, with flanking variablesequences on both sides of the motifs. Based on this obser-vation, we developed the ‘‘direct sequence’’ strategy to de-sign primer pairs from the variable sequences on both sidesof the microsatellite motifs. Thus, we applied two strategies,

    Fig. 6. Principal coordinate analysis of microsatellite loci showing relative positions of 172 individual trees of Acacia koa based on Dice’scoefficient. The three axes explain 40.3% (principal coordinate 1), 8.2% (principal coordinate 2), and 5.6% (principal coordinate 3) of thegenetic variation among the populations. The 11 populations were grouped into two major clusters, the Kauai populations forming one dis-tinct group and the populations from Oahu, Maui, and Hawaii forming the other group.

    Fredua-Agyeman et al. 1011

    # 2008 NRC Canada

  • ‘‘walking sequence’’ and ‘‘direct sequence’’, for the isolationof microsatellite loci from koa. From our results, the walk-ing sequence strategy produced more polymorphic loci thanthe direct sequence approach. The walking sequence strategyyielded an average of 6.5 alleles per locus, while the directsequence strategy produced 6.0 alleles per locus. The aver-age polymorphic information content (PIC) values for thewalking sequence strategy and direct sequence strategywere 0.51 and 0.42, respectively. Also, the walking se-quence strategy produced more primers with PIC values >0.6. There were also differences in the types of repeats thatwere isolated by the two strategies. Thirteen of 15 centralmotifs were different from the oligonucleotide repeats usedin the initial screening. This was because the initial PCRenrichment step selected fragments that contained microsa-tellite motifs at the two ends, while the central motifs wereindependent of this selection. Development of microsatellitemarkers by the direct sequence strategy is much easier andless expensive than by the walking sequence strategy. Fordeveloping microsatellite markers for a new species usingthe direct sequence strategy, we recommend screening atleast 100 clones obtained from the initial PCR amplificationusing oligonucleotide repeat primers. We developed 16markers from 96 clones that were obtained using 7 oligonu-cleotide repeat primers. Increasing the number of oligonu-cleotide repeat primers in the initial PCR step shouldfurther enhance the number of clones containing unique mi-crosatellite motifs. We observed that complementary pri-mers, such as (AC)10 and (TG)10, can lead to selection ofthe same clones.

    Although there are more than 1300 Acacia species distrib-uted in Australia, Africa, Asia, and North and South Amer-ica (Maslin et al. 2003; Miller et al. 2003; Murphy et al.2003), none of these have been characterized for genomicsequences. Microsatellite markers have so far been devel-oped for A. mangium, A. brevispica, A. saligna, and A. mel-lifera (Butcher et al. 2000; Otero-Arnaiz et al. 2005; Millarand Byrne 2007; Ruiz-Guajardo et al. 2007). The number ofmicrosatellite markers developed for the other acacias variesfrom 10 to 33. Millar and Byrne (2007) and Ruiz-Guajardoet al. (2007) isolated 10 and 11 microsatellite markers forA. saligna and A. mellifera, respectively, while Otero-Arnaizet al. (2005) and Butcher et al. (2000) developed 24 and 33

    markers to determine genetic diversity in A. brevispica andA. mangium, respectively. Microsatellite markers are largelyspecies-specific: among 33 markers developed for A. man-gium, only 4 could amplify the corresponding loci fromA. melanoxylon and none could amplify corresponding locifrom A. boliviana and A. nilotica (Butcher et al. 2000). Koais distinct from all other acacias and is found only in theHawaiian Islands. Koa is among the few tetraploid speciesof this genus (2n = 52); many other acacias, such A. mela-noxylon and A. mangium, are diploid (Elevitch et al. 2006).There are considerable phenotypic variations among koapopulations on different islands. The development of micro-satellite markers for koa will therefore facilitate future pop-ulation genetics studies and breeding programs for this tree.

    Classification of koa based on morphological characters isgreatly affected by the limited number of available pheno-typic characters and also by the ambiguities in the scoringof the phenotypic characters (Sun et al. 1996; Daehler et al.1999). Molecular markers have a unique power not pos-sessed by any other taxonomic method in allowing the esti-mation of similarity between organisms that are extremelydiverse (Kirst et al. 2005; Garcia-Martinez et al. 2006).Analyses of populations by molecular markers without thecomplementary morphological characters have been success-fully carried out in other plants (Staub et al. 2000; Palombiand Damiano 2002; Yifru et al. 2006). Both the UPGMAcluster analysis and the PCO subdivided the populationsinto two major groups: the group with the populations fromKauai was distinct from the group comprising the popula-tions of Oahu, Maui, and Hawaii. The UPGMA cluster anal-ysis further divided one of the major groups into threeadditional subgroups. Thus, among the koa populationsfrom the four major Hawaiian Islands, the Kauai populationshad the highest level of genetic diversity and formed adistinct group. This was not surprising considering thatKauai is the oldest among the major Hawaiian Islands (Mac-donald et al. 1983), permitting koa to evolve and differenti-ate into various ecological groups under a wide range ofmicro-environmental conditions. Also, Kauai possesses themost undisturbed forests among the islands sampled andhence its populations formed a distinct group. AlthoughHawaiian Islands are located far apart, the diversity in themicro-environments within each island appeared to play a

    Table 7. Pairwise genetic differentiation ðF0STÞ among 11 populations of Acacia koa based on 12 micro-satellite loci.

    LNR MRW MFR KFR NPC MKR HAL KHV HLR WNKBPH 0.070 0.151 0.078 0.074 0.123 0.124 0.070 0.147 0.129 0.184LNR — 0.212 0.132 0.112 0.129 0.135 0.091 0.160 0.103 0.097MRW — 0.098 0.070 0.166 0.135 0.132 0.173 0.144 0.230{

    MFR — 0.028* 0.094 0.057 0.105 0.101 0.087 0.145KFR — 0.079 0.036 0.086 0.071 0.063 0.115NPC — 0.117 0.077 0.175 0.131 0.171MKR — 0.085 0.099 0.091 0.131HAL — 0.166 0.124 0.154KHV — 0.032 0.120HLR — 0.028*WNK —

    *The lowest population differentiation detected among the populations analyzed.{The highest population differentiation detected among the populations analyzed.

    1012 Genome Vol. 51, 2008

    # 2008 NRC Canada

  • major role in the evolution of diversity in the koa popula-tion. If the environment within an island was homogeneous,the exchange of alleles between individual trees on the sameisland should have resulted in island ecotypes that were ge-netically more similar within islands than between islands.This phenomenon, referred to as the isolation-by-distance ef-fect, has been observed for other species (Epperson 2007;Medina et al. 2007; Bessert and Orti 2008; Worheide et al.2008). The calculated F 0ST statistics, used to compare the de-gree of heterozygosity, showed little evidence of populationdifferentiation among koa populations from the differentislands. Our findings that intra-island population diversitiescontributed more to the overall genetic diversity may havealso resulted in part from anthropogenic activities such asplanting of collected seeds during the last century.

    The microsatellite markers developed in this study will beuseful in koa breeding and improvement programs in thefuture. Many trees in koa plantations die owing to wilt dis-ease within a few years of planting (Whitesell 1990; Wilkin-son and Elvevitch 2003). In recent years, some koaprovenances have been identified that appear to be resistantto wilt disease (N. Dudley, Hawaii Agriculture ResearchCenter, Aiea, Hawaii, personal communication 2008). Insome crops, microsatellite loci linked to disease resistancehave been identified after analyzing a large number of sam-ples from disease-resistant and susceptible plants (Adhikariet al. 2004; Minamiyama et al. 2007; Bernet et al. 2008;Guo et al. 2008; Yang et al. 2008; Zhao et al. 2008). Micro-satellite loci linked to genes conferring resistance to wiltdisease may be identified and used as selectable markers infuture koa improvement programs. The markers developedin this study may also be used to determine whether theextent of genetic diversity in various koa plantations is sim-ilar to that found in natural koa forests. These markers mayalso be useful for distinguishing koa from other acacias.

    AcknowledgmentsThis research was supported by the USDA-TSTAR award

    No. 2005-34135-16007 and a McIntire-Stennis grant(560M). We would like to thank Candace Felling for helpwith koa sample collections and James Leary for field assis-tance; Sandro Jube, Kwesi Kutin, Ricelle Acob, and VishalNegi for laboratory assistance; and Aileen Yeh and NickDudley for technical support.

    ReferencesAdhikari, T.B., Wallwork, H., and Goodwin, S.B. 2004. Micro-

    satellite markers linked to the Stb2 and Stb3 genes for resistanceto Septoria tritici blotch in wheat. Crop Sci. 44: 1403–1411.

    Anderson, J.A., Churchill, G.A., Autrique, J.E., Tanksley, S.D., andSorrells, M.E. 1993. Optimizing parental selection for genetic-linkage maps. Genome, 36: 181–186. doi:10.1139/g93-024.PMID:18469981.

    Atchison, E. 1948. Studies on the Leguminoseae. II. Cytogeogra-phy of Acacia (Tourn.) L. Am. J. Bot. 35: 651–655. doi:10.2307/2438145.

    Bernet, G.P., Gorris, M.T., Carbonell, E.A., Cambra, M., andAsins, M.J. 2008. Citrus tristeza virus resistance in a core col-lection of sour orange based on a diversity study of three germ-plasm collections using QTL-linked markers. Plant Breed. 127:398–406. doi:10.1111/j.1439-0523.2008.01506.x.

    Bessert, M.L., and Orti, G. 2008. Genetic effects of habitat frag-mentation on blue sucker populations in the upper MissouriRiver (Cycleptus elongatus Lesueur, 1918). Conserv. Genet. 9:821–832. doi:10.1007/s10592-007-9401-4.

    Brookes, A.J. 1999. The essence of SNPs. Gene, 234: 177–186.doi:10.1016/S0378-1119(99)00219-X. PMID:10395891.

    Bussell, J.D., Waycott, M., and Chappill, J.A. 2005. Arbitrarily am-plified DNA markers as characters for phylogenetic inference.Perspect. Plant Ecol. Evol. Syst. 7: 3–26. doi:10.1016/j.ppees.2004.07.001.

    Butcher, P.A., Decroocq, S., Gray, Y., and Moran, G.F. 2000. De-velopment, inheritance and cross-species amplification of micro-satellite markers from Acacia mangium. Theor. Appl. Genet.101: 1282–1290. doi:10.1007/s001220051608.

    Carr, G.D. 1978. Chromosome numbers of Hawaiian floweringplants and the significance of cytology in selected taxa. Am. J.Bot. 65: 236–242. doi:10.2307/2442457.

    Coulaud, J., Brown, S.C., and Siljakyakovlev, S. 1995. First cyto-genetic investigation in populations of Acacia heterophylla, en-demic from La Réunion Island, with reference to A.melanoxylon. Ann. Bot. (Lond.), 75: 95–100. doi:10.1016/S0305-7364(05)80013-7.

    Daehler, C.C., Yorkston, M., Sun, W.G., and Dudley, N. 1999.Genetic variation in morphology and growth characters ofAcacia koa in the Hawaiian Islands. Int. J. Plant Sci. 160: 767–773. doi:10.1086/314163.

    De Silva, H.N., Hall, A.J., Rikkerink, E., McNeilage, M.A., andFraser, L.G. 2005. Estimation of allele frequencies in polyploidsunder certain patterns of inheritance. Heredity, 95: 327–334.doi:10.1038/sj.hdy.6800728. PMID:16094298.

    Dice, L.R. 1945. Measures of the amount of ecologic associationbetween species. Ecology, 26: 297–302. doi:10.2307/1932409.

    Elevitch, C.R., Wilkinson, K.M., and Friday, J.B. 2006. Acacia koa(koa) and Acacia koaia (koai‘a). In Species profiles for PacificIsland agroforestry. Edited by C.R. Elevitch. Permanent Agricul-ture Resources (PAR), Holualoa, Hawaii. pp. 1–29.

    Ellegren, H. 2004. Microsatellites: simple sequences with complexevolution. Nat. Rev. Genet. 5: 435–445. doi:10.1038/nrg1348.PMID:15153996.

    Epperson, B.K. 2007. Plant dispersal, neighbourhood size and iso-lation by distance. Mol. Ecol. 16: 3854–3865. doi:10.1111/j.1365-294X.2007.03434.x. PMID:17692081.

    Garcia-Martinez, S., Andreani, L., Garcia-Gusano, M., Geuna, F.,and Ruiz, J.J. 2006. Evaluation of amplified fragment lengthpolymorphism and simple sequence repeats for tomato germ-plasm fingerprinting: utility for grouping closely related tradi-tional cultivars. Genome, 49: 648–656. doi:10.1139/G06-016.PMID:16936844.

    Guo, Q., Zhang, Z.J., Xu, Y.B., Li, G.H., Feng, J., and Zhou, Y.2008. Quantitative trait loci for high-temperature adult-plantand slow-rusting resistance to Puccinia striiformis f. sp. triticiin wheat cultivars. Phytopathology, 98: 803–809. doi:10.1094/PHYTO-98-7-0803.

    Hamilton, J.A., and Eckert, C. 2007. Population genetic conse-quences of geographic disjunction: a prairie plant isolated onGreat Lakes alvars. Mol. Ecol. 16: 1649–1660. doi:10.1111/j.1365-294X.2007.03241.x. PMID:17402980.

    Hancock, J.M. 1995. The contribution of DNA slippage to eukaryo-tic nuclear 18S rRNA evolution. J. Mol. Evol. 40: 629–639.doi:10.1007/BF00160511. PMID:7643414.

    Jaccard, P. 1908. Nouvelles recherches sur la distribution flore.Bull. Soc. Vaud. Sci. Nat. 44: 223–270.

    Kirst, M., Cordeiro, C.M., Rezende, G.D.S.P., and Grattapaglia,D. 2005. Power of microsatellite markers for fingerprinting

    Fredua-Agyeman et al. 1013

    # 2008 NRC Canada

  • and parentage analysis in Eucalyptus grandis breeding popula-tions. J. Hered. 96: 161–166. doi:10.1093/jhered/esi023.PMID:15601907.

    Langer, A., and Koul, A.K. 1983. Studies on nucleolus and nucleo-lar chromosomes in angiosperms. 7. Nature of nucleolar chro-mosome polymorphism in Allium cepa var. viviparum (Metzg)Alef. Cytologia (Tokyo), 48: 323–332.

    Lian, C.L., Zhou, Z.H., and Hogetsu, T. 2001. A simple method fordeveloping microsatellite markers using amplified fragments ofinter-simple sequence repeat (ISSR). J. Plant Res. 114: 381–385. doi:10.1007/PL00014001.

    Litt, M., and Luty, J.A. 1989. A hypervariable microsatellite re-vealed by in vitro amplification of a dinucleotide repeat withinthe cardiac muscle actin gene. Am. J. Hum. Genet. 44: 397–401. PMID:2563634.

    Luo, Z.W., Hackett, C.A., Bradshaw, J.E., McNicol, J.W., andMilbourne, D. 2001. Construction of a genetic linkage map intetraploid species using molecular markers. Genetics, 157:1369–1385. PMID:11238421.

    Macdonald, G.A., Abbott, A.T., and Peterson, F.L. 1983. Volca-noes in the sea: the geology of Hawaii. University of HawaiiPress, Honolulu.

    Manifesto, M.M., Schlatter, A.R., Hopp, H.E., Suarez, E.Y., andDubcovsky, J. 2001. Quantitative evaluation of genetic diversityin wheat germplasm using molecular markers. Crop Sci. 41:682–690.

    Mantel, N. 1967. The detection of disease clustering and a general-ised regression approach. Cancer Res. 27: 209–220.PMID:6018555.

    Maslin, B.R., Miller, J.T., and Seigler, D.S. 2003. Overview of thegeneric status of Acacia (Leguminosae: Mimosoideae). Aust.Syst. Bot. 16: 1–18. doi:10.1071/SB02008.

    McCouch, S.R., Chen, X.L., Panaud, O., Temnykh, S., Xu, Y.B.,Cho, Y.G., et al. 1997. Microsatellite marker development,mapping and applications in rice genetics and breeding. PlantMol. Biol. 35: 89–99. doi:10.1023/A:1005711431474.PMID:9291963.

    Medina, A., Brethes, J.C., Sevigny, J.M., and Zakardjian, B. 2007.How geographic distance and depth drive ecological variabilityand isolation of demersal fish communities in an archipelagosystem (Cape Verde, Eastern Atlantic Ocean). Mar. Ecol.(Berl.), 28: 404–417. doi:10.1111/j.1439-0485.2007.00163.x.

    Millar, M.A., and Byrne, M. 2007. Characterization of polymorphicmicrosatellite DNA markers for Acacia saligna (Labill.) H.L.Wendl. (Mimosaceae). Mol. Ecol. Notes, 7: 1372–1374. doi:10.1111/j.1471-8286.2007.01890.x.

    Miller, J.T., Andrew, R., and Bayer, R.J. 2003. Molecular phyloge-netics of the Australian acacias of subg. Phyllodineae (Faba-ceae: Mimosoideae) based on the trnK intron. Aust. J. Bot. 51:167–177. doi:10.1071/BT01099.

    Minamiyama, Y., Tsuro, M., Kubo, T., and Hirai, M. 2007. QTLanalysis for resistance to Phytophthora capsici in pepper usinga high density SSR-based map. Breed. Sci. 57: 129–134. doi:10.1270/jsbbs.57.129.

    Morden, C.W., Caraway, V.C., and Motley, T.J. 1996. Develop-ment of a DNA library for native Hawaiian plants. Pac. Sci. 50:324–335.

    Murphy, D.J., Miller, J.T., Bayer, R.J., and Ladiges, P.Y.2003. Molecular phylogeny of Acacia subgenus Phyllodineae(Mimosoideae:Leguminosae) based on DNA sequences of theinternal transcribed spacer region. Aust. Syst. Bot. 16: 19–26.doi:10.1071/SB01042.

    Obbard, D.J., Harris, S.A., and Pannell, J.R. 2006a. Sexual systemsand population genetic structure in an annual plant: testing the

    metapopulation model. Am. Nat. 167: 354–366. doi:10.1086/499546. PMID:16673344.

    Obbard, D.J., Harris, S.A., and Pannell, J.R. 2006b. Simple allelic-phenotype diversity and differentiation statistics for allo-polyploids. Heredity, 97: 296–303. doi:10.1038/sj.hdy.6800862.PMID:16823402.

    Otero-Arnaiz, A., Schnabel, A., Glenn, T.C., Schable, N.A., Hagen,C., and Ndong, L. 2005. Isolation and characterization ofmicrosatellite markers in the East African tree, Acacia brevis-pica (Fabaceae: Mimosoideae). Mol. Ecol. Notes, 5: 366–368.doi:10.1111/j.1471-8286.2005.00929.x.

    Ouborg, N.J., Piquot, Y., and Van Groenendael, J.M. 1999. Popula-tion genetics, molecular markers and the study of dispersal inplants. J. Ecol. 87: 551–568. doi:10.1046/j.1365-2745.1999.00389.x.

    Palombi, M.A., and Damiano, C. 2002. Comparison betweenRAPD and SSR molecular markers in detecting genetic variationin kiwifruit (Actinidia deliciosa A. Chev). Plant Cell Rep. 20:1061–1066. doi:10.1007/s00299-001-0430-z.

    Randell, R.A., and Morden, C.W. 1999. Hawaiian Plant DNA li-brary II: endemic, indigenous, and introduced species. Pac. Sci.53: 401–417.

    Rozen, S., and Skaletsky, H.J. 2000. Primer3 on the WWW forgeneral users and for biologist programmers. In Bioinformaticsmethods and protocols: methods in molecular biology.Edited by S. Krawetz and S. Misener. Humana Press, Totowa,N.J. pp. 365–386.

    Ruiz-Guajardo, J.C., Otero-Arnaiz, A., Taylor, T., Stone, G.,Glenn, T.C., Schable, N.A., et al. 2007. Isolation of polymorphicmicrosatellite markers in the sub-Saharan tree, Acacia (Sene-galia) mellifera (Fabaceae: Mimosoideae). Mol. Ecol. Notes, 7:1138–1140. doi:10.1111/j.1471-8286.2007.01809.x.

    Schlötterer, C. 2004. The evolution of molecular markers — just amatter of fashion? Nat. Rev. Genet. 5: 63–69. doi:10.1038/nrg1249. PMID:14666112.

    Selkoe, K.A., and Toonen, R.J. 2006. Microsatellites for ecologists:a practical guide to using and evaluating microsatellite markers.Ecol. Lett. 9: 615–629. doi:10.1111/j.1461-0248.2006.00889.x.PMID:16643306.

    Shi, X. 2003. Genetic improvement of Leucaena and Acacia koaGray as high-value hardwoods. Ph.D. dissertation, Departmentof Tropical Plant and Soil Sciences, University of Hawaii atManoa, Honolulu.

    Siebert, P.D., Chenchik, A., Kellogg, D.E., Lukyanov, K.A., andLukyanov, S.A. 1995. An improved PCR method for walking inuncloned genomic DNA. Nucleic Acids Res. 23: 1087–1088.doi:10.1093/nar/23.6.1087. PMID:7731798.

    Skolnick, M.H., and White, R. 1982. Strategies for detecting andcharacterizing restriction fragment length polymorphisms(RFLP’s). Cytogenet. Cell Genet. 32: 58–67. doi:10.1159/000131687. PMID:6291862.

    St. John, H. 1979. Classification of Acacia koa and relatives (Legu-minosae). Pac. Sci. 33: 357–367.

    Staub, J.E., Danin-Poleg, Y., Fazio, G., Horejsi, T., Reis, N., andKatzir, N. 2000. Comparative analysis of cultivated melongroups (Cucumis melo L.) using random amplified polymorphicDNA and simple sequence repeat markers. Euphytica, 115: 225–241. doi:10.1023/A:1004054014174.

    Sun, W., Brewbaker, J.L., and Austin, M.T. 1996. Genetic variationof Acacia koa seed, seedling, and early growth traits. In Koa: adecade of growth. Proceedings of the Hawaii Forest IndustryAssociation Annual Symposium. Edited by D.O. Evans and L.Ferentinos. Honolulu. pp. 33–38.

    Tautz, D. 1989. Hypervariability of simple sequences as a general

    1014 Genome Vol. 51, 2008

    # 2008 NRC Canada

  • source for polymorphic DNA markers. Nucleic Acids Res. 17:6463–6471. doi:10.1093/nar/17.16.6463. PMID:2780284.

    Tautz, D., and Renz, M. 1984. Simple sequences are ubiquitous re-petitive components of eukaryotic genomes. Nucleic Acids Res.12: 4127–4138. doi:10.1093/nar/12.10.4127. PMID:6328411.

    Toth, G., Gaspari, Z., and Jurka, J. 2000. Microsatellites in differ-ent eukaryotic genomes: survey and analysis. Genome Res. 10:967–981. doi:10.1101/gr.10.7.967. PMID:10899146.

    Vos, P., Hogers, R., Bleeker, M., Reijans, M., Vandelee, T.,Hornes, M., et al. 1995. AFLP: a new technique for DNA fin-gerprinting. Nucleic Acids Res. 23: 4407–4414. doi:10.1093/nar/23.21.4407. PMID:7501463.

    Wang, Z.S., An, S.Q., Liu, H., Feng, J., Zhang, F., and Leng, X.2006. Effect of stand age and management regime on geneticdiversity of Thuidium cymbifolium in western China. Biol. Con-serv. 129: 551–557. doi:10.1016/j.biocon.2005.11.022.

    Weber, J.L. 1990. Informativeness of human (dC-dA)n�(dG-dT)npolymorphisms. Genomics, 7: 524–530. doi:10.1016/0888-7543(90)90195-Z. PMID:1974878.

    Weising, K., Atkinson, R.G., and Gardner, R.C. 1995. Genomicfingerprinting by microsatellite-primed PCR: a critical evalua-tion. PCR Methods Appl. 4: 249–255. PMID:7580910.

    Whitesell, C.D. 1990. Silvical characteristics of Acacia koa Gray.In Agriculture Handbook 654. Edited by H.R.M. Burns andB.H. Honkala. USDA Forest Service, Washington, D.C. pp. 17–28.

    Wilkinson, K.M., and Elvevitch, C.R. 2003. Growing koa: a Ha-waiian legacy tree. Permannet Agriculture Resources, Holualoa,Hawaii.

    Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A., andTingey, S.V. 1990. DNA polymorphisms amplified by arbitrary

    primers are useful as genetic markers. Nucleic Acids Res. 18:6531–6535. doi:10.1093/nar/18.22.6531. PMID:1979162.

    Worheide, G., Epp, L.S., and Macis, L. 2008. Deep genetic diver-gences among Indo-Pacific populations of the coral reef spongeLeucetta chagosensis (Leucettidae): founder effects, vicariance,or both? BMC Evol. Biol. 8: 24. doi:10.1186/1471-2148-8-24.PMID:18221552.

    Yadav, V.K., Kumar, S., and Panwar, R.K. 2007. Measurement ofgenetic dissimilarity in fieldpea (Pisum sativum L.) genotypesusing RAPD markers. Genet. Resour. Crop Evol. 54: 1285–1289. doi:10.1007/s10722-006-9109-1.

    Yang, C., Guo, W.Z., Li, G.Y., Gao, F., Lin, S.S., and Zhang, T.Z.2008. QTLs mapping for Verticillium wilt resistance at seedlingand maturity stages in Gossypium barbadense L. Plant Sci. 174:290–298. doi:10.1016/j.plantsci.2007.11.016.

    Yifru, T., Hammer, K., Huang, X.Q., and Roder, M.S. 2006. Regio-nal patterns of microsatellite diversity in Ethiopian tetraploidwheat accessions. Plant Breed. 125: 125–130. doi:10.1111/j.1439-0523.2006.01147.x.

    Zane, L., Bargelloni, L., and Patarnello, T. 2002. Strategies formicrosatellite isolation: a review. Mol. Ecol. 11: 1–16. doi:10.1046/j.0962-1083.2001.01418.x. PMID:11903900.

    Zhao, X.Q., Xu, J.L., Zhao, M., Lafitte, R., Zhu, L.H., Fu, B.Y., etal. 2008. QTLs affecting morph-physiological traits related todrought tolerance detected in overlapping introgression lines ofrice (Oryza sativa L.). Plant Sci. 174: 618–625. doi:10.1016/j.plantsci.2008.03.009.

    Zwettler, D., Vieira, C.P., and Schlotterer, C. 2002. Polymorphicmicrosatellites in Antirrhinum (Scrophulariaceae), a genus withlow levels of nuclear sequence variability. J. Hered. 93: 217–221. doi:10.1093/jhered/93.3.217. PMID:12195041.

    Fredua-Agyeman et al. 1015

    # 2008 NRC Canada

    /ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Average /GrayImageResolution 225 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.00000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Average /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.00000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects true /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName ( /PDFXTrapped /False

    /SyntheticBoldness 1.000000 /Description >>> setdistillerparams> setpagedevice